Detection of Abnormal Conditions Based on Analysis of Cyclic Movement in Scene Image

Abstract

Automatic detection of abnormal conditions in scene images is an important research theme. This paper proposes a general-purposed method for automatic detection of abnormal conditions based on the analysis of periodicity of movements. In the proposed method, cyclic movements are detected by voting binarized averaged grayscale values into each frequency. The disorder of movements, which represents the occurrence of abnormal situation, can be recognized by monitoring the change of bias conditions in voting. The experimental results showed that the reliable detections for periodicity and the speedy detections of abnormal conditions were realized in the general situations.